Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

AI changed what high performance looks like

Why are we still measuring it the same way?

[Photo: Getty Images]

Profile photo of Doug Dennerline

CEO · Betterworks

One statistic from Betterworks’ latest research has stuck with me. Of the human resources leaders we surveyed, 90% say AI has changed what a high performer looks like. That’s a remarkable level of agreement. In business, it’s rare to find 90% of leaders aligned on anything.

But the statistic raises a bigger question. If AI changed what high performance looks like, why do most organizations still measure performance the same way as before AI became part of everyday work?

We’ve spent years talking about how to modernize performance management. Annual reviews became quarterly reviews, followed by quarterly reviews that became continuous feedback. Organizations invested in new processes, new technologies and new ways to collect employee feedback.

Those changes mattered, but I think we’re focused on the wrong problem. The bigger challenge is that many of the signals we’ve historically used to identify high performers are becoming less reliable because AI is changing how performance is created.

That’s why so many organizations are struggling to answer a deceptively simple question: What does high performance actually look like now?

Output no longer offers the full story

For most of my career, output was a pretty reliable signal. If someone consistently produced more, delivered stronger results, or had a bigger impact than their peers, it was usually safe to assume they were performing at a higher level.

Today, that relationship is becoming more complicated as AI changes the scale of what one person can accomplish. No matter the industry, we’re seeing employees produce at levels that would have been difficult to imagine just a few years ago. A marketer can create and test campaigns faster while a salesperson can personalize outreach at a scale that wasn’t practical before. The result looks more like adding the output of five or ten people together.

While this is exciting, it also makes performance harder to understand. When one person can suddenly generate the output that once required a larger team, leaders naturally want to know what’s driving those results.

Often, it’s some combination of expertise, experience, judgment, and yes, even AI. And that’s where traditional performance signals start to fall short. For years, output and contribution were closely linked. If someone produced more value, it usually showed up in their work. AI is separating those two things.

The highest performers are distinguished by how they use AI to create value.

We still measure individuals in a human-AI world

Many organizations get stuck on how performance management was built around the simple assumption that work is performed by people. Human resource departments would measure the individuals’ contribution and reward them based on results.

That model made sense when most work was created entirely by humans, but not as humans increasingly lean into AI. AI drafts, analyzes, summarizes, recommends, and automates, with humans refining the output by providing context, judgment, creativity, and prioritizing what matters.

The final outcome increasingly reflects both human judgment and AI capability, with the highest performers working effectively alongside AI. Can they identify when AI is wrong? Can they apply judgment when the answer isn’t obvious? Can they decide which work should be automated and which work still requires a human touch?

Those capabilities don’t always show up in a performance review, but they increasingly determine who creates the most value.

Why traditional performance signals break down

I believe many organizations are starting to feel a disconnect between how work happens and how performance gets measured. Two employees can produce similar outputs while contributing very different levels of value.

One employee might accept an AI-generated recommendation without much thought. But another might challenge it, improve it, identify risks, and arrive at a better outcome. While those contributions can look remarkably similar, they’re anything but.

It’s the same with managers. Historically, if a team wasn’t performing, managers looked at people, priorities, or processes. AI introduces another variable, and as it becomes part of everyday work, understanding performance requires more context, as signals become less reliable on their own.

What leaders should measure instead?

Leaders need new signals, and I believe that four signals matter more than ever.

  • Judgment: The ability to know when AI is right, when it’s wrong, and when a situation requires human experience, context or nuance.
  • Adaptability: How quickly someone can learn new tools, adjust to changing priorities, and continue creating value as work evolves.
  • Orchestration: The ability to determine which work should be done by people, which work should be done by AI, and how various combinations produce better outcomes than either could alone.
  • Impact: The ability to turn the above into meaningful business results. While activity and productivity matter, organizations only succeed when work drives outcomes.

These signals emerge over time, through the decisions employees make, the way they use technology, and the outcomes they create. They’re not captured in an annual review or a single performance rating.

The opportunity ahead

AI is changing what great performance looks like. That means performance management can no longer rely on the same signals it always has. Successful organizations will recognize and reward the capabilities AI can’t replace, including sound judgment, continuous learning, and the ability to translate technology into meaningful business impact.

Getting performance right is a competitive advantage that will shape who attracts, develops, and retains the talent needed to lead in the AI era.

Doug Dennerline is CEO at Betterworks.